Rapid and unplanned urbanization has ushered in a variety of public health challenges, including exposure to traffic pollution and greater dependence on automobiles. Moreover, vulnerable population groups often bear the brunt of negative outcomes and are subject to disproportionate exposure and health effects. This makes it imperative for urban transportation engineers, land use planners, and public health professionals to work synergistically to understand both the relationship between urban design and population exposure to traffic pollution, and its social distribution. Researchers have started to pay close attention to this connection, mainly by conducting observational studies on the relationship between transportation, urban form, and air quality. However, research on this topic is still nascent. Further, most studies do not predict exposures under alternative urban design scenarios. Hence, to understand the relationship between urban design and population exposures, there is a need to build and apply integrated modeling tools that can predict exposures under alternative urban design scenarios. Within this context, the overarching goal of this dissertation is to understand how the transportation infrastructure of cities can be designed for improved urban air quality and mitigation of population exposure to traffic pollution. The study area is Hillsborough County, Florida, a sprawling region with limited transit availability and a diverse population along with a mix of urban, suburban, and rural areas. The rank of the county for sprawl and congestion metrics (i.e., yearly delay and travel time index) fall in the mid-range in comparison with other US urban regions. Thus, the study area may be representative of other US urban regions with medium sprawl and above-average congestion levels. Oxides of nitrogen (NOx), a surrogate for traffic pollution, is the focus pollutant. The Health Effects Institute’s report on traffic-related air pollution identifies NOx as a potential surrogate due to its relative ease of measurement and the abundance of epidemiologic studies that characterize exposures to NOx. Because exposures are dependent on the spatial and temporal distributions of both people and pollution, this study first sought to understand the importance of activity and travel patterns of individuals for exposure estimation. To estimate exposures, the 2009 National Household Travel Survey (NHTS) data containing daily individual activity records, ArcGIS-estimated shortest-time travel route profiles, and the annual-average diurnal cycle of NOx derived from hourly CALPUFF dispersion model results from 2002, were combined. Two exposure measures were estimated: activity-based exposure that considers the daily activity and travel patterns of individuals, and residence-based exposure that considers only the pollutant concentrations at the residences. Exposure estimation without inclusion of activity and travel patterns was found to slightly underestimate activity-based exposures on average. Additionally, disproportionately-high exposures were found for blacks, Hispanics, below poverty groups, urban residents, and people whose daily travel time is greater than one hour. Finally, urbanicity and travel time variables were found to be the strongest predictors of daily exposure. Following this, a modeling framework was developed to predict population exposure by integrating activity-based travel demand modeling (DaySim), dynamic traffic assignment simulation (MATSim), mobile-source emission estimation (EPA MOVES), and pollutant dispersion modeling (R LINE). This modeling framework was used to predict daily population and subgroup exposures by estimating the high-resolution spatial and temporal distributions of both pollution and individual activities for the year 2010. Persistent exposure inequalities were found at the population-level; blacks, Hispanics, active age groups (19-65 years), below-poverty and middle-income groups, urban residents, and individuals with daily travel times above one hour had higher estimated exposures than the population mean. These inequalities for blacks, Hispanics, and below-poverty non-white groups worsened at higher exposure levels. Use of low-resolution activity and pollution data as opposed to high-resolution data led to underestimation of exposures (by 10% on average). Finally, the integrated modeling framework was employed to understand the relationship between urban transportation and land use design, air quality, and population exposure. Three scenarios that are based on a combination of diesel-bus transit services and residential distribution were simulated. Specifically, the low-transit scenario used the 2040 base residential distribution and the 2010 bus services. The enhanced-transit scenario applied the 2040 bus services proposed for the county instead. The compact-growth scenario added an increase of residential density to this latter scenario. Specifically, about 37% of total households were redistributed from locations with low accessibility to jobs and transit to locations near employment and bus stops. Results indicate slight higher non-car travel mode shares in the enhanced-transit and compact-growth scenarios compared to the low-transit scenario (with a 7.1% increase for walking, 0.2% for bicycle, and 1.8% for transit for the compact-growth scenario versus the low-transit scenario). The enhanced-transit scenario resulted in slightly lower daily total travel distances and times compared with the low-transit scenario, but daily total emissions and winter mean concentration of NOx were higher, i.e., the increase in bus transit services did not induce sufficient shifts in travel mode to overcome the concomitant increase in diesel-bus emissions. The compact-growth scenario resulted in lower daily total travel distance (9%) and travel time (2.1%) and daily total emissions of NOx (11%) and its winter mean concentration (9%), compared with both the low-transit and enhanced-transit scenarios. Although the compact-growth scenario improved the air quality of the region on average, daily population mean exposure was higher compared with both the low-transit (29%) and enhanced-transit scenarios (25%). This is largely due to the redistribution of population to urban core locations that had higher pollutant levels. Overall, neither the bus-transit improvements nor residential compaction strategies alone were sufficient to mitigate population exposures. Combining them with transit that services both origins and destinations, uses clean fuel technologies, and separates major roadways from dense residential pockets may be needed for greater exposure reductions. Overall, this dissertation has implications for population exposure to traffic pollution and public health through transportation and land use interventions. Results presented here may be applicable to other study regions that have similar composite sprawl scores as the Tampa Bay area. Future studies should exploit spatially-and temporally-resolved data on human activities and travel, vehicular activities, and air quality for better characterization of population exposure. Engineers and planners should pay greater attention to integrated land use and transport planning; lone, disjointed, and ill-planned design interventions may exacerbate population exposure to air pollution. The integrated modeling framework presented here may be applied in a wide variety of urban contexts to further explore the nexus between travel demand, air quality, and exposures. However, before such an exercise is undertaken, a preliminary analysis should be conducted to assess the transferability of the framework. Policies that could be studied include mixed land use design, urban compaction with controlled sociodemographic distributions (to assess exposure inequality), and inclusion of additional types of transit and fuel technologies.